TargetBC

Exploring policy options to reduce emissions of black carbon and other superpollutants

by developing an emulator tool to inform policy-making on the role of black carbon in climate change 

The TargetBC project is funded by the Clean Air Fund under its 'superpollutants' initiative focussing on black carbon, and is being implemented between May 2025 and May 2027

Project partners (U.Aarhus, Denmark; University of Leeds/UK Meteorological Office, UK; CICERO, Norway; NILU, Norway; University of Stockholm, Sweden; AMAP Secretariat)  and associated partners (Environment and Climate Change Canada and University of Montreal, Quebec)

Aim and main activities: The project work  involves performing fixed-SST and fully-coupled Earth System Model (ESM) simulations, the results of which will be used to 'train' an emulator tool that can be used to investigate black carbon emissions reduction options to inform policy-making on climate responses. The (OpenSource) emulator development builds on a version applied in the AMAP 2021 SLCF assessment work, but with substantial improvements including the ability to run on the basis of emissions rather than air concentrations, and training the model using multi-model ESM results from five state-of-the-art ESMs (GISS-E2.1, NorESM2.0.6, UKESM1.1, CanESM5.1-PAM, and CESM2). 

Additional project components include inverse modelling using the FLEXPART CTM and observational data to validate source-receptor relationships.

The regional focus of the work is on the Arctic with a wider/global application on emission sources in Europe and Asia (India and China). 

TargetBC ESM simulations

TargetBC: Description of ESM experiments

Task: Delivering ESM simulation results to underpin the development and 'training' of the TargetBC emulator

Authors: Marianne T. Lund, Ulas Im, Cynthia Whaley, Steven Turnock, Zhihong Zhuo, Peter Langen;

Version: April 2026

As part of the TargetBC project, a set of ESM experiments have been designed, perturbing emissions of black carbon (BC) and sulfur dioxide (SO2) in different latitude bands and geographical regions. The following describes the experiment setup and output. This document also contains directions for accessing, using, and acknowledging the data (Section 4).

In addition to a year 2014 control simulation, simulations where emissions are perturbed individually in 4 broad latitude bands and 4 geographical regions were performed.  An additional experiment perturbing global shipping emissions is requested. Table 1 summarizes the experiments. Simulations have been performed with 5 models: GISS-E2.1, NorESM2.0.6, UKESM1.1, CanESM5.1-PAM, and CESM2. A brief overview of these models is provided in Appendix 1, including information about their treatment of aerosols and deposition of BC on snow. 

1.Experiment details: 

  • Participating models were used in their CMIP6 configuration and all emissions and forcing data were as per CMIP6
  • For the control and latitude band perturbations, both fixed SST (free running AMIP style) and ocean coupled simulations were requested
  • For the geographical region perturbations only fixed SST simulations were requested
  • Fixed SST simulations were run for between 30 and 45 years, using the last 20 for the estimation of  effective radiative forcing (ERF)
  • Coupled simulations were run for 100 years, discarding as spin-up a model dependent number of years (see Table 2) 
  • For all experiments, one realization/ensemble member was run 
  • Models used their own default prescribed sea surface temperature or fSST runs (either year 2014 or climatology – see individual model details in Table 2)
  • Models did not use a common restart file 

2. Emission perturbations were implemented as follows:

  • All perturbations were 10x year 2014  CMIP6 emissions of BC and SO2, respectively, keeping everything else fixed in the control simulation configuration. [One model (GISS-E2.1) used a climatological mean of the biomass burning emissions as the basis for the 10x perturbation, the rest perturbed single year emissions. A preliminary analysis showed that 2014 did not stand out as an extreme biomass burning year, at least globally. ]
  • For black carbon: 
    -  Latitude bands: all experiments are performed with the perturbation applied to land anthropogenic (i.e. no international shipping or aviation) plus biomass burning emissions  AND with the perturbation applied to land anthropogenic emissions only.
    - Geographical regions: the perturbation applied to only land anthropogenic 
  • For sulfur dioxide: 
    - Latitude bands: perturbation applied to land anthropogenic emissions only. 
    - Geographical regions: perturbation applied to land anthropogenic emissions only
  • The definition of the geographical regions follows those used in the HTAP3-OPNS experiments. The region mask file is available here.   

Notes:

(1) One model (GISS-E2.1) used a climatological mean of the biomass burning emissions as the basis for the 10x perturbation, the rest perturbed single year emissions. A preliminary analysis showed that 2014 did not stand out as an extreme biomass burning year, at least globally. 

(2) Due to a scaling factor within the model being applied to CMIP6 configuration, he UKESM simulations are actually x20 biomass burning (emissions data should reflect this). 

3. Model output request and file naming convention: 

The models were requested to output standard CMIP6 monthly mean variables. A complete list of requested variables is given in Appendix 2. 
For models who can do this, selected additional variables at daily resolution were requested: 
ta, trefht, prect, psl, fluxes (rsut, rsdt, rlut, rsds, rsus, rlds, rlus), concbc, od550aer, abs550aer, cl, clt

The naming of output files roughly follows the CMIP6 convention: 
<variable>_<timeres>_<modelname>_<experimentname>_<experiment_type> [optional:_<realization>_<model_grid>_<time_period>].nc
variable = CMIP6 name. timeres = mon or day. experiment: experiment short name from Table 1, experiment type: atm for fSST runs, cpl for coupled ocean runs. 

4. Data access and further use: 
Model output is stored in part on the Electronic Research Data Archive at Aarhus University (ERDA), with additional variables available from individual institutions storage facilities. Data will be made openly available once documented but is at present available for further analysis upon request. If using one or more data sets in a publication, we ask that you offer co-authorship to the people who designed and carried out the simulations. Contact points and people to include in further publications are for each model:
-    CanESM5.1-PAM: Cynthia Whaley 
-    GISS-E2.1: Ulas Im
-    NorESM2: Marianne T. Lund, Maura Dewey 
-    UKESM1.1: Steven Turnock 
-    CESM2: Zhihong Zhuo 
For questions related to ERDA: Ulas Im. 

A further OpenAccess archive will be established for model results datasets in connection with planned publications presenting project results.

We kindly ask that you acknowledge the TargetBC project in any use of these data, as follows: “This dataset was produced as part of the Clean Air Fund (CAF) project TargetBC”. 
 

Table 1: Summary of requested experiments

Experiment short names for output file naming (see Section 3) where glb=global, arc=Arctic (65-90N), nhm=northern hemisphere midlatitudes (15-65N), trp=tropics (15S-15N) and she=Southern Hemisphere (15-90S). Ant=anthropogenic land-based emissions; lnd=anthropogenic land based + biomass burning emissions; shp= shipping emissions; lsa=land anthropogenic + shipping; all= land anthropogenic + biomass burning + shipping.

Experiment full name
Experiment short name
Region of perturbation
Emissions perturbed
fSST and coupled

2014 control

2014CTRL

n/a

n/a

Y

Black Carbon

--

--

--

--

Global Anth 10xBC

glbBCant

Global

Land anthropogenic

Y

Global Anth+BB 10xBC

glbBClnd

Global

Land anthropogenic + biomass burning (BB)

Y

Ocean 10xBC

glbBCshp

Global

Shipping

Y

Arctic Anth+BB 10xBC

arcBClnd

65N - 90N

Land anthropogenic + BB

Y

NH mid-lat Anth+BB 10xBC

nhmBClnd

15N - 65N

Land anthropogenic + BB

Y

Tropics Anth+BB 10xBC

trpBClnd

15S - 15N

Land anthropogenic + BB

Y

SH Anth+BB 10xBC

sheBClnd

15S - 90S

Land anthropogenic + BB

Y

Arctic Anth 10xBC

arcBCant

65N - 90N

Land anthropogenic

fSST only

NH mid-lat Anth 10xBC

nhmBCant

15N - 65N

Land anthropogenic

fSST only

Tropics Anth 10xBC

trpBCant

15S - 15N

Land anthropogenic

fSST only

SH Anth 10xBC

sheBCant

15S - 90S

Land anthropogenic

fSST only

East Asia Anth 10xBC

easBCant

HTAP EAS

Land anthropogenic

fSST only

South Asia Anth 10xBC

sasBCant

HTAP SAS

Land anthropogenic

fSST only

North America Anth 10xBC

namBCant

HTAP NAM

Land anthropogenic

fSST only

Europe Anth 10xBC

eurBCant

HTAP EMEP_West

Land anthropogenic

fSST only

Sulfur dioxide

--

--

--

--

Global Anth 10xSO2

glbSO2ant

Global

Land anthropogenic

Y

Global Anth + ship 10xSO2

glbSO2lsa

Global

Land anthropogenic + shipping

Y

Ocean 10xSO2

glbSO2shp

Global

Shipping

Y

Global Anth+BB 10xSO2

glbSO2all

Global

Land anthropogenic + shipping

fSST only

Arctic 10xSO2

arcSN2ant

65N - 90N

Land anthropogenic

Y

NH mid lat 10xSO2

nhmSO2ant

15N - 65N

Land anthropogenic

Y

Tropics Anth 10xSO2

trpSO2ant

15S - 15N

Land anthropogenic

Y

SH 10xSO2

sheSO2ant

15S - 90S

Land anthropogenic

Y

East Asia 10xSO2

easSO2ant

HTAP EAS

Land anthropogenic

fSST only

South Asia 10xSO2

sasSO2ant

HTAP SAS

Land anthropogenic

fSST only

North America 10xSO2

namSO2ant

HTAP NAM

Land anthropogenic

fSST only

Europe 10xSO2

eurSO2ant

HTAP EMEP_West

Land anthropogenic

fSST only

Table 2: model-specific information about runs

Model Name
SST information – 2014 or climatology
fSST run total length and spin-up
Coupled run total length and spin-up

NorESM2.0.6

2014

BC: 30 years, 10 years spin-up. SO2: 45 years, 15 years spin-up

100 years for SO2, 200 years for BC. (nc files contain all years; data prior to equilibration of global surface air temperature should be considered spin up (number of years depend on experiment)).

GISS-E2.1

2014

45 years, 15 years spin-up

100 years (71 years spin up)

UKESM1.1

2014

45 years, 15 years spin-up

100 years (70 years spin up)

CanESM5.1-PAM

2014

45 years, 15 years spin-up (nc files contain only the last 30 years)

100-120 years (depending on the perturbation), 70-90 years spin-up (nc files contain only the last 31 years)

CESM2

2014

45 years, 15 years spin-up

tbc

Appendix 1: Information about participating models

Model
Type
Horizontal grid
Vertical grid
Aerosol Scheme
Treatment of BC on snow

GISS-E2.1

ESM

2° × 2.5° lat/long grid

40 levels, up to 0.1 hPa

MATRIX (Bauer et al., 2008), which is a microphysical aerosol model, is used, which can track the mixing state of several aeroesol populations. DMS, sea-salt, mineral dust, and biogenic emissions are calculated onlline.

Hansen and Nazarenko (1994)

NorESM2.0.6

ESM

1.9° × 2.5° lat/long grid

32 levels, lid ~3 hPa, hybrid sigmapressure

CAM6-Nor (based on CAM6 (Danabasoglu et al, 2020)) with OsloAero aerosol scheme (Seland et al., 2020; Kirkevåg et al., 2013; 2018). Modal scheme simulating aerosol mass and number.

SNICAR (Snow, Ice, and Aerosol Radiative model) (Flanner et al. 2007)

UKESM1.1

ESM

145 × 192 (1.875° × 1.25° lat/ long grid)

85 (hybrid height terrain following grid) levels, up to ~0.01 hPa

Aerosol scheme is GLOMAP-Mode (Modal scheme, simulating aerosol mass and number). Separate mass-based bin scheme used for minerals dust. SO4, BC, OM, sea salt in 5 log-normal modes (nucleation to coarse) and dust in 6 size segregated bins. Online emissions of DMS, sea-salt and dust aerosols, as well as emissions of primary marine organics and biogenic organic compounds.
Archibald et al., (2020), Mulcahy et al., (2020), Mulcahy et al., (2023)

Currently no representation of BC deposition on snow.

CanESM5.1-PAM

ESM

128x64, T63 (~2.8∘ × 2.8∘)

49 vertical hybrid levels extending up to 1 hPa

Piecewise lognormal approximation Aerosol Model (PAM) (von Salzen et al, 2013)

Namazi et al (2015)

CESM2

ESM

0.9° lat x 1.25° lon

70 vertical hybrid levels extending up to ~6 × 10⁻⁶ hPa

Modal Aerosol Model version 4 (MAM4; Liu et al., 2014)

SNICAR (Snow, Ice, and Aerosol Radiative model), (Flanner et al. 2007)

Appendix 2: List of output variables

Table/dimensions
Name
Description
Units
Note

Core variables

Amon

(month,lat, lon)

rsdt

TOA Incident Shortwave Radiation

W m-2

rsut

TOA Outgoing Shortwave Radiation

W m-2

rlut

TOA Outgoing Longwave Radiation

W m-2

Rsutcs

TOA Outgoing Clear-Sky Shortwave Radiation

W m-2

Rlutcs

TOA Outgoing Clear-Sky Longwave Radiation

W m-2

Hfls

Surface Upward Latent Heat Flux

W m-2

Hfss

Surface Upward Sensible Heat Flux

W m-2

Rsds

Surface Downwelling Shortwave Radiation

W m-2

Rsus

Surface Upwelling Shortwave Radiation

W m-2

Rlds

Surface Downwelling Longwave Radiation

W m-2

Rlus

Surface Upwelling Longwave Radiation

W m-2

Rsdscs

Surface Downwelling Clear-Sky Shortwave Radiation

W m-2

Rsuscs

Surface Upwelling Clear-Sky Shortwave Radiation

W m-2

Rldscs

Surface Downwelling Clear-Sky Longwave Radiation

W m-2

Tas

Near-Surface Air Temperature

K

Ts

Surface Temperature

K

Clt

Total Cloud Cover Percentage

%

Psl

Sea Level Pressure

Pa

(month, lev, lat, lon)

Pfull

Pressure at Model Full-Levels

Pa

Limon

(month,lat, lon)

Snc

Snow Area Percentage

%

Snd

Snow Depth

m

AERmon

(month,lat, lon)

drybc

Dry Deposition Rate of Black Carbon Aerosol Mass

kg m-2 s-1

dryso2

Dry Deposition Rate of SO2

kg m-2 s-1

wetbc

Wet Deposition Rate of Black Carbon Aerosol Mass

kg m-2 s-1

wetso2

Wet Deposition Rate of SO2

kg m-2 s-1

emibc

Total Emission Rate of Black Carbon Aerosol Mass

kg m-2 s-1

emiso2

Total Emission Rate of SO2

kg m-2 s-1

(month, lev, lat, lon)

mmrbc

Elemental Carbon Mass Mixing Ratio

kg kg-1

so2

SO2 Volume Mixing Ratio

mol mol-1

airmass

Vertically Integrated Mass Content of Air in Layer

kg m-2

mmrso4

Aerosol Sulfate Mass Mixing Ratio

kg kg-1

Additional variables requested

Amon

(month,lat, lon)

pr

Precipitation rate

m/s

ps

Surface pressure

Pa

rsdt

TOA Incident Shortwave Radiation, double call

W m-2

For IRF calculations, i.e. a call to the radiation scheme without aerosols present. Incident would be the same

rsutaf

TOA Outgoing Shortwave Radiation, double call

W m-2

For IRF calculations, i.e. a call to the radiation scheme without aerosols present

rlutaf

TOA Outgoing Longwave Radiation, double call

W m-2

For IRF calculations, i.e. a call to the radiation scheme without aerosols present

(month, lev, lat, lon)

hus

Specific humidity

kg/kg

ua

Zonal wind

m/s

va

Meridional wind

m/s

wap

Omega (Vertical air pressure tendency)

Pa s-1

pfull

Pressure

Pa

zg

Geopotential height

m

ta

Air temperature

K

AERmon

(month,lat, lon)

aod550abs

Absorbtion AOD

Unitless

aod550

AOD

Unitless

day

(lon,lat,day)

tas

Surface temperature

K

tasmax

Maximum surface temperature

K

tasmin

Minimum surface temperature

K

pr

Precipitation rate

m/s

References

Archibald et al. 2020 https://doi.org/10.5194/gmd-13-1223-2020 

Bauer et al. 2008 https://doi.org/10.5194/acp-8-6003-2008 

Danabasoglu et al, 2020 https://doi.org/10.1029/2019MS001916 

Flanner et al. 2007 https://doi.org/10.1029/2006JD008003 

Hansen and Nazarenko 1994 https://doi.org/10.1073/pnas.2237157100

Kirkevåg et al. 2018 https://doi.org/10.5194/gmd-11-3945-2018 

Liu et al. 2016 https://doi.org/10.5194/gmd-9-505-2016

Mulcahy et al. 2020 https://doi.org/10.5194/gmd-13-6383-2020 

Mulcahy et al. 2023 https://doi.org/10.5194/gmd-16-1569-2023 

Namazi et al. (2015) https://doi.org/10.5194/acp-15-10887-2015 

von Salzen et al. 2013  https://doi.org/10.1080/07055900.2012.755610